Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Prepared in cooperation with the Colorado Department of Public Health and Environment

Scientific Investigations Report 2010–5037

U.S. Department of the Interior U.S. Geological Survey Cover. Pueblo Reservoir in south-central Colorado. Photograph by Robert Stogner. Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

By M. Alisa Mast and David P. Krabbenhoft

Prepared in cooperation with the Colorado Department of Public Health and Environment

Scientific Investigations Report 2010–5037

U.S. Department of the Interior U.S. Geological Survey U.S. Department of the Interior KEN SALAZAR, Secretary

U.S. Geological Survey Marcia K. McNutt, Director

U.S. Geological Survey, Reston, Virginia: 2010

For more information on the USGS—the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment, visit http://www.usgs.gov or call 1-888-ASK-USGS For an overview of USGS information products, including maps, imagery, and publications, visit http://www.usgs.gov/pubprod To order this and other USGS information products, visit http://store.usgs.gov

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Although this report is in the public domain, permission must be secured from the individual copyright owners to reproduce any copyrighted materials contained within this report.

Suggested citation: Mast, M.A., and Krabbenhoft, D.P., 2010, Comparison of mercury in water, bottom sediment, and zooplankton in two Front Range reservoirs in Colorado, 2008–09: U.S. Geological Survey Scientific Investigations Report 2010–5037, 20 p. iii

Contents

Abstract...... 1 Introduction...... 1 Purpose and Scope...... 2 Description of Study Area...... 2 Summary of Fish-Tissue Mercury in Colorado...... 4 Methods of Investigation...... 4 Field Sampling...... 4 Sediment-Incubation Experiments...... 6 Analytical Methods...... 7 Quality Control...... 7 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs...... 8 Chemical Characteristics...... 8 Mercury in Surface Water...... 10 Mercury in Reservoir Bottom Sediment...... 11 Mercury in Zooplankton...... 13 Sediment-Incubation Experiments...... 13 Factors Affecting Bioaccumulation of Mercury...... 15 Summary...... 18 References...... 18

Figures

1. Maps showing location of study areas and sampling sites in (A) Pueblo Reservoir and (B) Brush Hollow Reservoir, and of (C) reservoirs with walleye (Stizostedion vitreum) sampled during the Colorado Department of Public Health and Environment fish-tissue mercury study...... 3 2–9. Graphs showing: 2. Average monthly storage in Pueblo Reservoir (1974–2008) and Brush Hollow Reservoir (1960–2008) for the period of record...... 4 3. Mercury concentrations in fish tissue from selected water bodies in Colorado (A) for fish species with detectable concentrations and (B) for walleye (Stizostedion vitreum)...... 5 4. Dissolved oxygen and temperature profiles during 2008 and 2009 at site 7B in Pueblo Reservoir and site S1 in Brush Hollow Reservoir...... 9 5. Relation of methylmercury and the ratio of methylmercury to total mercury concentrations with water depth for reservoir-bottom sediments collected during August 2008 in Pueblo and Brush Hollow Reservoirs...... 12 6. Relation between total mercury concentrations and methylmercury concentrations in zooplankton samples collected from Pueblo and Brush Hollow Reservoirs during August 2008...... 13 7. Methylmercury concentration relative to duration of the experiment in days for the sediment-incubation experiments...... 14 iv

8. Relation between maximum methylmercury concentration in water and cores in sediment-incubation experiments...... 14 9. Percent change in reservoir surface area between annual minimum and maximum water level in Pueblo and Brush Hollow Reservoirs during 2002–2004...... 16 10. Photographs showing (A) submerged macrophytes and (B) grass-covered littoral zone at north end of Brush Hollow Reservoir in late summer 2008, and (C–D) typical rocky shoreline at Pueblo Reservoir...... 17

Tables

1. Sampling sites at Pueblo and Brush Hollow Reservoirs, Colorado...... 6 2. Quality-control results for water samples collected during the study...... 7 3. Quality-control results for solid samples collected during the study...... 8 4. Physical properties and dissolved constituent concentrations in water samples from Pueblo and Brush Hollow Reservoirs, Colorado...... 10 5. Mercury concentrations in water samples from Pueblo and Brush Hollow Reservoirs, Colorado...... 11 6. Mercury concentrations in reservoir-bottom sediments from Pueblo and Brush Hollow Reservoirs, Colorado...... 12 v

Conversion Factors

Multiply By To obtain Length inch (in.) 2.54 centimeter (cm) foot (ft) 0.3048 meter (m) mile (mi) 1.609 kilometer (km) Area acre 0.004047 square kilometer (km2) Volume ounce, fluid (fl. oz) 0.02957 liter (L) gallon (gal) 3.785 liter (L) acre-foot (acre-ft) 1,233 cubic meter (m3) Mass ounce, avoirdupois (oz) 28.35 gram (g) pound, avoirdupois (lb) 0.4536 kilogram (kg)

Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows: °F=(1.8×°C)+32 Concentrations of chemical constituents in water are given in milligrams per liter (mg/L) or nanograms per liter (ng/L). Mercury concentrations in sediment and zooplankton are given in nanograms per gram (ng/g) dry weight and concentrations in fish are reported in micrograms per gram (μg/g) wet weight.

Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

By M. Alisa Mast and David P. Krabbenhoft

Abstract mercury concentrations between the reservoirs. Due to the shallow depth and the large annual water-level fluctuations at Brush Hollow Reservoir, proportionally larger areas of The U.S. Geological Survey, in cooperation with the shoreline at Brush Hollow Reservoir are subjected to annual Colorado Department of Public Health and Environment, reflooding compared to Pueblo Reservoir. Moreover, presence conducted a study to investigate environmental factors that of macrophyte beds and regrowth of terrestrial vegetation may contribute to the bioaccumulation of mercury in two likely increase the organic content of near-shore sediments in Front Range reservoirs. One of the reservoirs, Brush Hollow Brush Hollow Reservoir, which may stimulate methylmercury Reservoir, currently (2009) has a fish-consumption advisory production in littoral areas subject to reflooding. Results of for mercury in walleye (Stizostedion vitreum), and the other, a laboratory incubation experiment were consistent with this Pueblo Reservoir, which is nearby, does not. Water, bottom hypothesis. sediment, and zooplankton samples were collected during 2008 and 2009, and a sediment-incubation experiment was conducted in 2009. Total mercury concentrations were low in midlake water samples and were not substantially differ- Introduction ent between the two reservoirs. The only water samples with detectable methylmercury were collected in shallow areas of Mercury is released to the environment primarily from Brush Hollow Reservoir during spring. Mercury concentra- anthropogenic sources (burning of fossil fuels and wastes) and tions in reservoir bottom sediments were similar to those is transported to most aquatic ecosystems through atmospheric reported for stream sediments from unmined basins across pathways (U.S. Environmental Protection Agency, 1997). the United States. Despite higher concentrations of fish-tissue Mercury is of concern because it bioaccumulates in fish and mercury in Brush Hollow Reservoir, concentrations of methyl- can pose health risks to humans and wildlife that consume mercury in sediment were as much as 3 times higher in Pueblo large amounts of contaminated fish. As of 2008, 48 States in Reservoir. Mercury concentrations in zooplankton were at the the United States have issued fish-consumption advisories for low end of concentrations reported for temperate lakes in the mercury and 26 States have issued statewide advisories (http:// Northeastern United States and were similar between sites, epa.gov/waterscience/fish/advisories/). Nearly all mercury in which may reflect the seasonal timing of sampling. fish occurs as methylmercury, which is a more bioavailable Factors affecting bioaccumulation of mercury were and toxic form of mercury that appears to be produced largely assessed, including mercury sources, water quality, and by sulfate-reducing bacteria in anoxic environments (Gilmour reservoir characteristics. Atmospheric deposition was deter- and others, 1992). The extent to which methylmercury bioac- mined to be the dominant source of mercury; however, due to cumulates in fish depends not only on the nature and length the proximity of the reservoirs, atmospheric inputs likely are of the food chain but also on landscape and water-chemistry similar in both study areas. Water-quality constituents com- characteristics (Driscoll and others, 2007). Greater bioac- monly associated with elevated concentrations of mercury in cumulation of methylmercury has been found to occur in oli- fish (pH, alkalinity, sulfate, nutrients, and dissolved organic gotrophic lakes that have low pH and low alkalinity (Wiener carbon) did not appear to explain differences in fish-tissue and others, 2003; Chen and others, 2005) and in lakes whose mercury concentrations between the reservoirs. Low methyl- watersheds are dominated by runoff from wetlands (St. Louis mercury concentrations in hypolimnetic water indicate low and others, 1994; Kolka and others, 1999). Lakes that undergo potential for increased methylmercury production following seasonal stratification can contain elevated concentrations of the development of anoxic conditions in summer. Based on methylmercury following the development of anoxic condi- the limited dataset, water-level fluctuations and shoreline tions in the hypolimnion (Driscoll and others, 1995; Watras characteristics appear to best explain differences in fish-tissue and others, 2005). Some studies have reported lower mercury 2 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09 accumulation in nutrient-rich water bodies due to biodilu- criterion for the State of Colorado in some tissue samples from tion of mercury under algal-bloom conditions (Pickhardt and Brush Hollow Reservoir. others, 2002). Mercury in fish commonly is elevated in man- made reservoirs compared to natural lakes. In newly created reservoirs, the initial flooding of organic-rich soils results in Description of Study Area elevated mercury concentrations in fish for 10 to 20 years after Pueblo Reservoir is approximately 3 miles northwest of the reservoir is created (Bodaly and others, 2007). Mercury Pueblo, Colorado (fig. 1A), and has a total storage capacity of accumulation also appears to be elevated in established reser- 357,678 acre-feet (Bureau of Reclamation, 1977). The reservoirs that experience annual water-level fluctuations related voir, which began filling in 1974, was created to store water to water storage, power generation, or flood control (Sorensen for municipal, industrial, and irrigation uses but also pro- and others, 2005) and natural lakes that experience large vides flood control, recreational activities, sport fishing, and water-level fluctuations related to climate (Selch and others, wildlife habitat for the region (Lewis and Edelmann, 1994). 2007). At full pool, the reservoir is about 9 miles long and ranges in In 2004, the Colorado Department of Public Health width from 0.3 to about 2.2 miles and has a maximum depth and Environment (CDPHE) initiated a 5-year study to test of 155 feet near the dam (Bureau of Reclamation, 1972). fish tissue (fillets) for mercury in water bodies across the Nearly all the inflow to the reservoir is from the Arkansas State (http://www.cdphe.state.co.us/wq/fishcon/index.html, River, and more than one-half of the inflow occurs during May accessed January 2010). Of the 94 water bodies tested as of through July (Lewis and Edelmann, 1994). Annual storage August 2008, nearly one-quarter had one or more fish species typically peaks in April then decreases through the summer that exceeded the Colorado tissue criterion for mercury of and early autumn months because of decreased inflow and 0.5 microgram per gram wet weight (µg/g ww) for the protec- large downstream demands for irrigation water (fig. 2). The tion of human health (http://www.cdphe.state.co.us/wq/fishcon/ reservoir inundates four large canyons and parts of several index.html, accessed January 2010). As a result of the tissue small tributaries. The shoreline is irregular and rocky and the study and a growing public awareness of mercury contami- reservoir is underlain by flat-lying shales, sandstones, and nation in Colorado, the U.S. Geological Survey (USGS), in limestones (Galloway and others, 2008). Vegetation is sparse cooperation with the CDPHE, conducted a study to investigate with scattered willows, grasses, and cottonwoods growing in environmental factors that might contribute to the bioaccu- ravines and on small sandy beaches. The reservoir supports mulation of mercury in fish. Two reservoirs in Colorado were both cold- and warm-water fisheries and is routinely stocked selected for the study—one that has a current (2009) fish- with rainbow trout, cutbow trout, walleye, bass, wiper, and consumption advisory for mercury in walleye (Stizostedion channel catfish http://wildlife.state.co.us/Fishing/Reports/( vitreum) and one that does not. Because of the large number FisherySurveySummaries/, accessed January 2010). of lakes and reservoirs in the State, the results of this study Brush Hollow Reservoir is 6 miles northeast of Florence, may be useful to resource managers in identifying untested Colorado (fig. 1B), and has a storage capacity of 3,933 acre- water bodies that are at greatest risk for mercury contamina- feet. The reservoir was established in 1907 to store irrigation tion. Additionally, insight gained from this study also may be water from the Beaver Creek drainage to the east, which is useful in the development of management strategies to lower delivered by way of a supply ditch that discharges near the methylmercury production in Colorado reservoirs and reduce dam. On rare occasions, rainstorms upstream from the reser- mercury available to fish. voir contribute some water. The reservoir is within the Brush Hollow Wildlife Refuge and also provides recreation, sport Purpose and Scope fishing, and wildlife habitat. At full pool, the reservoir is about 0.9 mile long and 0.3 mile wide and has a maximum depth The purpose of this report is to compare mercury concen- of 45 feet near the dam. The reservoir is subject to extreme trations in water, bottom sediment, and zooplankton samples annual fluctuations in water level due to water management collected from two Front Range reservoirs in Colorado during activities and in many years is drawn down to the minimum 2008 and 2009 and to evaluate factors that may be contribut- allowable pool of 25 feet near the dam (D. Krieger, Colorado ing to bioaccumulation of mercury in fish. Additionally, this Division of Wildlife, oral commun., 2008). Annual storage report presents results of a sediment-incubation experiment typically peaks in April around 4,000 acre-feet then declines conducted in 2009 to evaluate the methylation potential rapidly, reaching a minimum of less than 1,000 acre-feet by of sediments in these two reservoirs. The two reservoirs, midsummer (fig. 2). The reservoir is situated on flat-lying Brush Hollow and Pueblo Reservoirs (fig. 1), are located in shale and limestone units (Scott and others, 1978), and veg- the Arkansas River drainage within 20 miles of each other etation around the reservoir is dominated by cottonwoods, yet have very different fish-tissue mercury concentrations. grasses, and sagebrush. Beds of aquatic macrophytes grow in Mercury concentrations in walleye, which is the top predator shallow areas along the northern end of the reservoir when it fish in each reservoir, are below analytical detection in tissue is at full pool. The reservoir is managed as a warm-water and samples from Pueblo Reservoir but exceeded the fish-tissue seasonal trout fishery and is routinely stocked with rainbow Introduction 3

C Reservoirs with walleye 1. Bonny Reservoir 2. Boyd Reservoir 3. Brush Hollow Reservoir COLORADO 18! 4. Carter lake 8 ! 2 !!! 7 9 5. Chatfield Reservoir B 4 ! ! !11 6. Cherry Creek Reservoir 20 7. Douglas Reservoir 16 6 1 8. Horsetooth Reservoir ! 5 ! ! ! Front Range 9. Jackson Reservoir Metropolitan 10. Juniata Reservoir 10 Lead ville 11. Lake Loveland ! }Area 12. John Martin Reservoir 13. Narraguinnep Reservoir BH2 3 ! 14 14. Nee Grande Reservoir !( ! ! 12 15. Pueblo Reservoir 15 ! 16. Rifle Gap Reservoir Study area Arkansas River 17. Sanchez Reservoir 13 18. Smith #2 Reservoir ! 21 S5 ! 19. Trinidad Reservoir 17 19 20. Union Reservoir !( ! ! !( S4 21. Vallecito Reservoir !( !( 105°05′ 104°49′30″ BH1 BH3 !( Brush 0 1 2 3 4 MILES S3 Hollow Reservoir 0 1 2 3 4 KILOMETERS Beaver S7!( !( !( Creek S2 S8

Brush Hollow Reservoir 38°23′30″ Florence S1!( Arkansas S6

!( D

i Dam t c h River

0 0.1 0.2 0.3 0.4 MILE Pueblo Reservoir

0 0.1 0.2 0.3 0.4 KILOMETER

Pueblo

PR1 !( !( !( 3C PR2 EXPLANATION !( !( Pueblo Reservoir ! Incubation core PR3 4B ! Sediment ! Sediment and water

!( D ! Water !( a

!( m 5C 7B Supply ditch 6C City

Base from U.S. Geological Survey digital data, 1:100,000 0 1 2 3 4 MILES Universal Transverse Mercator projection Zone 12 0 1 2 3 4 KILOMETERS

Figure 1. Location of study areas and sampling sites in (A) Pueblo Reservoir and (B) Brush Hollow Reservoir, and of (C) reservoirs with walleye (Stizostedion vitreum) sampled during the Colorado Department of Public Health and Environment fish-tissue mercury study. 4 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

250,000 Pueblo 4,000 200,000

3,000 150,000

Brush Hollow 2,000 IN ACRE-FEET 100,000

1,000 STORAGE, IN ACRE-FEET 50,000 BRUSH HOLLOW RESERVOIR PUEBLO RESERVOIR STORAGE,

0 0 OCT DEC FEB APR JUNE AUG MONTH

Figure 2. Average monthly storage in Pueblo Reservoir (1974–2008) and Brush Hollow Reservoir (1960–2008) for the period of record. trout, cutbow trout, Snake River cutthroat, black crappie, blue- piscivorous species. For example, in the 21 reservoirs with gill, channel catfish, and walleye http://wildlife.state.co.us/( walleye (figs. 1 and 3B), nearly one-half had walleye tissue Fishing/Reports/FisherySurveySummaries/, accessed January concentrations near or below analytical detection while the 2010). remainder had elevated concentrations, which in two reservoirs exceeded 0.8 µg/g ww. For the two reservoirs in this study, mercury in walleye was below detection in all tissue Summary of Fish-Tissue Mercury in Colorado samples collected from Pueblo Reservoir but was detected in all tissue samples collected from Brush Hollow Reservoir, As part of the CDPHE fish-tissue mercury study, sam- with concentrations ranging from 0.14 to 0.57 µg/g ww. ples were collected from 94 water bodies across the State as of August 2008, representing 38 different species of fish. Concentration data and collection methods for the CDPHE study are available online at http://www.cdphe.state.co.us/wq/ Methods of Investigation fishcon/index.html. The most frequently collected fish species were walleye (Stizostedion vitreum; 17 percent), rain- This section of the report describes methods of field sam- bow trout (Oncorhynchus mykiss; 10 percent), smallmouth pling, the approach used for the sediment-incubation experi- bass (Micropterus dolomieu; 7 percent), northern pike (Esox ments, and laboratory analytical methods. Mercury results lucius; 6 percent), black crappie (Pomoxis nigromaculatus; for quality-control samples (blanks and replicates) collected 6 percent), brown trout (Salmo trutta; 5 percent), largemouth during the study also are presented. Sampling sites at the two bass (Micropterus salmonoides; 5 percent), wiper (Morone reservoirs are listed in table 1 and shown in figure 1. saxatilis; 5 percent), channel catfish Ictalurus( punctatus; 4 percent), and green sunfish Lepomis( cyanellus; 3 percent). Field Sampling Mercury concentrations were above analytical detection limits (0.1 µg/g ww) in 38 percent of the 3,767 fish-tissue Water samples for mercury were collected at each res- samples analyzed. Detected concentrations ranged from 0.1 ervoir using a Teflon Kemmerer sampler that was precleaned to 1.5 µg/g ww with a median concentration of 0.2 µg/g ww. with 5-percent hydrochloric acid (HCl) and deionized water. Mercury concentrations varied considerably among species, Unfiltered samples were transferred to precleaned Teflon with the highest concentrations in the piscivorous species bottles and acidified in the field to 1 percent HCl by volume. such as northern pike, walleye, wiper, and bass and the lowest During August 2008, water samples for mercury were col- concentrations in forage fish such as rainbow and brook trout lected from the epilimnion and hypolimnion at two locations (fig. 3A). Mercury concentrations exceeded the U.S. Envi- in Pueblo Reservoir (sites 7B and 4B) and one location in ronmental Protection Agency criterion of 0.3 µg/g ww for Brush Hollow Reservoir (site S1) (fig. 1). One equipment protection of human health (http://www.epa.gov/waterscience/ blank was collected from the Kemmerer sampler and one field criteria/) in 11 percent of samples and exceeded the State of replicate was collected at each reservoir. Samples also were Colorado tissue criterion level of 0.5 µg/g ww in 4 percent of collected with the Kemmerer at all three sites for major dis- samples. For individual fish species, mercury concentrations solved constituents, nutrients, and dissolved organic carbon; also varied considerably among water bodies, particularly for these samples were field filtered through a 0.45-micron (µm) Methods of Investigation 5

(37) (18) (28) (30) (26 (19) (49) (50) (84) (57) (35) (136) (142) (113) (333) (161) 1.5

1.0

0.5 FISH -TISSUE MERCURY, -TISSUE MERCURY, FISH IN MICROGRAM PER GRAM WET WEIGHT GRAM PER IN MICROGRAM

EXPLANATION (12) Number of values 0.0 Outlier 1.5 times quartile range

WIPER SPLAKE 75th percentile BLUEGILL WALLEYE MACKINAW BROOK TROUT WHITE SUCKER COMMON CARP BROWN TROUT BLACK CRAPPIEYELLOW PERCH NORTHERN PIKE RAINBOW TROUT CHANNEL CATFISH SMALLMOUTHLARGEMOUTH BASS BASS Median

FISH SPECIES 25th percentile B Colorado fish-consumption criterion 0.5 microgram per gram, wet weight 1.5 (45) (18) (60) (15) (24) (60) (16) (71) (39) (10) (10) (16) (21) (11) (60) (8) (12) (7) (80) (11) (12) U.S. Environmental Protection Agency fish-consumption criterion 0.3 microgram per gram, wet weight Analytical detection limit 0.1 microgram per gram 1.0

0.5 FISH -TISSUE MERCURY IN WALLEYE, -TISSUE MERCURY FISH IN MICROGRAM PER GRAM WET WEIGHT GRAM PER IN MICROGRAM

0.0

BOYD BONNY CARTER PUEBLO UNION JUNIATA SMITH #2 CHATFIELD DOUGLAS JACKSON LOVELAND RIFLE GAPSANCHEZ TRINIDAD VALLECITO HORSETOOTH NEE GRANDE CHERRY CREEK JOHN MARTIN BRUSH HOLLOW NARRAGUINNEP WATER BODY

Figure 3. Mercury concentrations in fish tissue from selected water bodies in Colorado A( ) for fish species with detectable concentrations and (B) for walleye (Stizostedion vitreum). [U.S. Environmental Protection Agency (USEPA) fish-consumption criterion = 0.3 microgram per gram, wet weight; Colorado fish-consumption criterion = 0.5 microgram per gram, wet weight; Analytical detection limit = 0.1 microgram per gram] 6 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09 capsule filter, packed on ice, and transported to the laboratory. sample mass for the analysis. Two samples were collected in At each site a Secchi depth was recorded and field measure- the deepest part of each reservoir and two were collected in ments of water temperature, specific conductance, and dis- shallower water. The zooplankton samples were drained of solved oxygen were made at depth intervals of 2–3 feet using most water and transferred to precleaned Teflon vials. Sedi- a Hydrolab water-quality sonde. During 2009, additional water ment and zooplankton samples were packed on ice and trans- samples and field replicates for mercury were collected during ported to the processing facility where they were frozen and April, May, and June. At midlake sites, samples were collected shipped to the laboratory for mercury analysis. using the Kemmerer sampler, and at sites S6–S8, grab samples from the surface were collected into Teflon bottles. Bottom sediments were collected from five sites in each Sediment-Incubation Experiments reservoir in August 2008 approximately along a transect from the upstream to downstream end (fig. 1). At the time For the incubation experiments, exposed littoral sedi- of sampling, the reservoir level at Brush Hollow Reservoir ments were cored from each reservoir at three different loca- was declining and site S5 was at the northern shoreline of the tions during spring of 2009 (fig. 1). A column of sediment reservoir. Sediments were collected using a benthos gravity approximately 2 inches thick and 5 inches in diameter was corer fitted with a 2.5-inch-diameter polyethylene core tube. excavated at each sampling site by using a polyvinylchloride The cores were extruded in the field, and the top 0.75 inch of (PVC) ring and a ceramic knife. Care was taken not to disturb sediment was removed from the core using a plastic spatula the sediment surface while the core was placed surface-side and placed in a double polyethylene bag and homogenized. At up in the bottom of a 2-liter Teflon jar. At the time of sedi- site S4 in Brush Hollow Reservoir the surface-sediment sec- ment collection, a sample of reservoir water was collected tion was split and one-half was submitted as the environmental near the shore and filtered into a 6-liter Teflon container using sample and one-half as a field replicate. Four zooplankton an acid-rinsed 0.45-µm capsule filter. The sediment cores and samples were collected at each reservoir (table 1) during water were packed on ice and transported to the laboratory the August 2008 sampling event by using a zooplankton net and stored at 4°C until the start of the experiment. A spike of (150-μm mesh) fitted with a Teflon cod-end assembly. Hori- isotopically enriched inorganic mercury-201 was added to the zontal tows behind the boat were required to obtain enough filtered water sample and allowed to mix for 24 hours. The

Table 1. Sampling sites at Pueblo and Brush Hollow Reservoirs, Colorado.

[No., site number used in fig. 1; USGS, U.S. Geological Survey]

No. Site name USGS station number Sample medium 7B Pueblo Reservoir Site 7B 381602104435200 Water, sediment, zooplankton. 6C Pueblo Reservoir Site 6C 381548104453300 Sediment. 5C Pueblo Reservoir Site 5C 381559104465500 Sediment. 4B Pueblo Reservoir Site 4B 381647104475300 Water, sediment, zooplankton. 3C Pueblo Reservoir Site 3C 381729104494100 Sediment. PR1 Pueblo sediment core 1 381737104494501 Incubation core. PR2 Pueblo sediment core 2 381726104485501 Incubation core. PR3 Pueblo sediment core 3 381703104482401 Incubation core.

S1 Brush Hollow Reservoir Site S1 382734105030601 Water, sediment, zooplankton. S2 Brush Hollow Reservoir Site S2 382744105030601 Sediment. S3 Brush Hollow Reservoir Site S3 382753105030601 Sediment, zooplankton. S4 Brush Hollow Reservoir Site S4 382800105030501 Sediment. S5 Brush Hollow Reservoir Site S5 382802105030501 Sediment. S6 Brush Hollow Ditch 382729105030001 Water. S7 Brush Hollow Reservoir nr Boat Ramp 382747105025801 Water. S8 Brush Hollow Creek at Hwy 50 382522105031701 Water. BH1 Brush Hollow sediment core 1 382758105030801 Incubation core. BH2 Brush Hollow sediment core 2 382809105030601 Incubation core. BH3 Brush Hollow sediment core 3 382758105030001 Incubation core. Methods of Investigation 7 isotopic spike, which resulted in an initial 2 ng/L concentra- liter (ng/L) for total mercury and 0.04 ng/L for methylmercury tion of mercury-201, was added to monitor the production of in water. Reservoir bottom sediment and zooplankton samples methylmercury during the experiments. Each 2-liter Teflon jar were digested with acid and analyzed for total mercury and was filled with the spiked water solution, which was added methylmercury using the method described for water, with slowly to minimize disturbance of sediment at the bottom of some modifications (DeWild and others, 2004; Olund and the jar. The jars were closed with a screw-top lid and allowed others, 2004). The method detection limits in sediment and to incubate at room temperature in the laboratory. Three zooplankton are 2.0 nanograms per gram (ng/g) for total samples were collected from each jar during the experiment at mercury and 0.02 ng/g for methylmercury. Major constituents 2- to 3-week intervals. During sampling, the jars were opened and dissolved organic carbon in surface water were analyzed briefly and a small volume of water was removed using Teflon at a Colorado Water Science Center research laboratory, and tubing and a peristaltic pump. The samples were filtered in nutrients were analyzed at the USGS National Water Quality a vacuum chamber, using a 0.45-µm quartz fiber-filter, into Laboratory using published analytical methods (Fishman and precleaned Teflon bottles and preserved to 1 percent HCl by Friedman, 1989). Percent organic carbon in sediments was volume. determined by loss on ignition at the USGS Mercury Research Laboratory (Fishman and Friedman, 1989). Data for all field samples are stored in the USGS National Water Informa- Analytical Methods tion System (NWIS) at http://waterdata.usgs.gov/nwis under USGS station numbers listed in table 1. All samples were analyzed for total mercury and methyl mercury at the USGS Mercury Research Laboratory in Middleton, Wisconsin. Total mercury in water (both unfiltered Quality Control and filtered samples) was analyzed by cold-vapor atomic fluorescence spectroscopy (CVAFS) using EPA Method 1631 The quality of the mercury analyses was evaluated Revision E (U.S. Environmental Protection Agency, 2002). through collection of blank and replicate samples. Two water Methylmercury in water (both unfiltered and filtered samples) blanks were collected during the study, one in the field using was analyzed using the method described by DeWild and the Kemmerer sampler and one in the laboratory (for methyl- others (2001) except that detection was by inductively coupled mercury only) using the filtration chamber (table 2). Neither plasma/mass spectrometry with isotope dilution instead of by sample had detectable concentrations, indicating that the CVAFS (T. Sabin, U.S. Geological Survey, written commun., potential for mercury contamination in water samples result- 2009). The method detection limits are 0.4 nanogram per ing from sample collection, processing, and analysis was low.

Table 2. Quality-control results for water samples collected during the study.

[ng/L, nanogram per liter; MDL, method detection limit; <, less than; --, not measured; RPD, relative percent difference; %, percent]

Methylmercury, ng/L Total mercury, ng/L (MDL = 0.04) (MDL = 0.40) Field blank <0.04 <0.04 Laboratory filter blank <.04 -- Pueblo Reservoir Site 7B Environmental <0.04 0.53 Replicate <.04 .54 RPD -- 2% Environmental <.04 .41 Replicate <.04 .45 RPD -- 9% Brush Hollow Reservoir Site S1 Environmental <0.04 0.60 Replicate <.04 .64 RPD -- 6% Environmental .050 .60 Replicate .058 .56 RPD 15% 7% 8 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Four field replicates were collected at surface-water sites, Reservoir during late summer (Lewis and Edelmann, 1994). At two during 2008 and two during 2009 (table 2). The relative the shallower site 4B in Pueblo Reservoir, dissolved oxygen percent difference in concentration (calculated as the dif- concentration also decreased with depth during the August ference of the sample pairs divided by the mean of the pairs sampling although concentrations did not fall below 4.0 mg/L, multiplied by 100) was 15 percent for the one sample with likely due to more efficient vertical mixing in the upstream detectable methylmercury and less than 10 percent for total reaches of the reservoir (Lewis and Edelmann, 1994). Pueblo mercury indicating good analytical precision for mercury in Reservoir typically mixes during October (Lewis and Edel- water. One field replicate was collected for sediment and one mann, 1994) and remains relatively well mixed through April; for zooplankton during 2008 (table 3). The relative percent in early summer the water temperature stratifies and dissolved difference in these solid samples was higher for methylmer- oxygen levels begin a seasonal decline in the hypolimnion. cury (13 percent for sediment and 12 percent for zooplankton) During late summer (August 2008) at site S1 in Brush Hol- than it was for total mercury (1 percent for both sediment and low Reservoir, dissolved oxygen concentration decreased zooplankton), but overall the results were within the expected to 1 mg/L but thermal stratification was minimal due to the precision for the analytical methods used. shallow depth of the reservoir (fig. 4). The June 2009 profile indicates hypoxic conditions develop slightly earlier in Brush Hollow Reservoir than in Pueblo Reservoir. Major dissolved constituents and nutrients measured in Comparison of Mercury in Water, water from the two reservoirs during August 2008 are shown Bottom Sediment, and Zooplankton in in table 4. Specific conductance, which is proportional to the concentration of major dissolved constituents (Hem, 1985), Two Front Range Reservoirs was slightly higher in Brush Hollow Reservoir than in Pueblo Reservoir. Calcium was the dominant cation in both reservoirs. The following sections of the report describe the chemi- Bicarbonate, represented as alkalinity in table 4, was the domi- cal characteristics of Pueblo and Brush Hollow Reservoirs and nant anion in Pueblo Reservoir and bicarbonate and sulfate present total mercury and methylmercury concentration data were codominant in Brush Hollow Reservoir. Silica concentra- for water, bottom sediment, and zooplankton samples col- tions were substantially lower in Pueblo Reservoir, particularly lected during 2008 and 2009. Results of a sediment-incubation in the epilimnion at site 7B than in Brush Hollow Reservoir. experiment conducted during 2009 also are presented. The depletion of silica in the epilimnion likely reflects the large diatom population that is present in the reservoir during the summer (Lewis and Edelmann, 1994). Dissolved organic Chemical Characteristics carbon (DOC) concentrations, which commonly are correlated with mercury in fish (Chen and others, 2005), were slightly Profiles of dissolved oxygen and temperature at site 7B higher in Brush Hollow Reservoir than in Pueblo Reservoir, in Pueblo Reservoir and site S1 in Brush Hollow Reservoir but overall concentrations in both reservoirs were fairly low are depicted in figure 4. In both reservoirs, dissolved oxygen (less than 5.0 mg/L). Nitrate and phosphorus species were near and water temperature decreased with increasing water depth or below detection levels in all but the deepest samples from during August 2008. At site7B in Pueblo Reservoir, dissolved Pueblo Reservoir owing to high biological uptake in the mid- oxygen concentration was 0.3 milligram per liter (mg/L) at dle of summer. Total dissolved nitrogen, which was composed a depth of 105 feet. This is consistent with a previous study, primarily of organic nitrogen, was as much as 2 times higher which reported that dissolved oxygen concentrations less than in Brush Hollow Reservoir than Pueblo Reservoir, indicating 1 mg/L commonly occurred in the deepest area of Pueblo Brush Hollow is the more productive of the two reservoirs.

Table 3. Quality-control results for solid samples collected during the study.

[MDL, method detection limit; Env., environmental sample; Rep., replicate sample; RPD, relative percent difference; %, percent; concentrations in nanogram per gram dry weight]

Sediment Zooplankton Constituent MDL Env. Rep. RPD Env. Rep. RPD Methylmercury 0.02 0.14 0.16 13% 13.2 11.7 12% Total mercury 2.0 24.6 24.9 1% 19.8 19.7 1% Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs 9

The productivity or trophic state of the reservoirs can be reservoir is slightly more productive. This perhaps is due to directly compared using the trophic state index (TSI), which shallower depths and warmer water temperatures in Brush was computed based on secchi-disk depth using the method of Hollow Reservoir, which can stimulate aquatic plant growth. Carlson (1977). For Pueblo Reservoir, the TSI ranged from 50 A higher trophic level for Brush Hollow Reservoir also is indi- to 55, indicating the reservoir is borderline between meso- cated by the macrophyte beds that grow around the perimeter trophic and eutrophic conditions during summer (table 4). of the reservoir during early summer but are largely absent For Brush Hollow Reservoir, the TSI was 59 indicating the from Pueblo Reservoir.

August 2008 April 2009 June 2009 DISSOLVED OXYGEN CONCENTRATION, IN MILLIGRAMS PER LITER 1 3 5 7 9 1 3 5 7 9 1 3 5 7 9 0 Site 7B Site 7B Site 7B

60 WATER DEPTH, IN FEET WATER

100

8 12 16 20 24 0 Site S1 Site S1

EXPLANATION Dissolved oxygen Temperature 20 WATER DEPTH, IN FEET WATER

8 12 16 20 24 8 12 16 20 24 WATER TEMPERATURE, IN DEGREES CELSIUS

Figure 4. Dissolved oxygen and temperature profiles during 2008 and 2009 at site 7B in Pueblo Reservoir and site S1 in Brush Hollow Reservoir. 10 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Table 4. Physical properties and dissolved constituent concentrations in water samples from Pueblo and Brush Hollow Reservoirs, Colorado.

[--, not applicable; °C, degress Celsius; μS/cm, microsiemens per centimeter at 25°C; TSI = 60 - 14.41 x ln (secchi depth x 0.3048); Concentrations in milligrams per liter; CaCO3, calcium carbonate; N, nitrogen; <, less than; P, phosphorus]

Pueblo Reservoir Pueblo Reservoir Brush Hollow Reservoir Site 7B Site 4B Site S1 Property or constituent 08/05/2008 08/05/2008 08/27/2008 Epilimnion Hypolimnion Epilimnion Hypolimnion Epilimnion Hypolimnion Sampling depth, feet 5 90 5 28 3 17 Secchi depth, feet 6.5 -- 4.5 -- 3.5 -- Water temperature, °C 24.7 18.2 24.4 23.1 21.5 20.7 Specific conductance, μS/cm 282 242 273 286 369 373 Oxygen, dissolved 7.6 1.3 8.0 4.8 7.8 3.4 pH, standard units 8.52 8.11 8.61 8.30 8.69 8.29 Trophic state index (TSI) 50 -- 55 -- 59 --

Calcium, dissolved 33.4 30.7 32.8 33.5 40.2 41.1 Magnesium, dissolved 8.1 6.3 7.9 8.1 7.9 7.9 Sodium, dissolved 10.7 7.6 10.5 10.9 22.7 22.8 Potassium, dissolved 1.7 1.4 1.7 1.7 2.5 2.5 Chloride, dissolved 4.1 2.7 3.8 3.9 5.0 5.0 Sulfate, dissolved 54.2 36.7 50.8 51.0 86.8 85.7 Silica, dissolved .4 6.7 1.5 2.5 10.0 10.6

Alkalinity as CaCO3 73.8 72.6 72.8 74.5 81.9 83.1 Organic carbon, dissolved 3.0 2.8 2.7 2.3 4.1 4.1 Nitrate as N, dissolved <.006 .252 <.006 .021 <.006 <.006 Total nitrogen as N, dissolved .236 .393 .140 .196 .796 .423 Phosphorus as P, dissolved <.006 .014 .008 .006 .006 .005 Orthophosphate as P, dissolved .003 .014 .004 .004 <.006 <.006

Mercury in Surface Water due to settling of particles from the overlying water column. During spring of 2009, concentrations of methylmercury and The range of methylmercury and total mercury concentra- total mercury at the midlake sites at both reservoirs (7B and tions measured in surface-water samples collected during the S1) were low, although methylmercury concentrations are not study are shown in table 5. Concentrations of methylmercury available for the Brush Hollow Reservoir samples collected are particularly important because they are a key indicator of in April 2009. Low methylmercury concentrations indicate potential mercury exposure to aquatic organisms. Sample dis- very limited methylmercury production in the open water of tribution differs somewhat between the two reservoirs because both reservoirs, even during late summer when strong anoxic of the short duration and limited scope of this study as well as conditions develop in the hypolimnion. some missing methylmercury analyses, and the results need Several additional water samples were collected from to be interpreted with some caution. In August 2008, methyl- sites in and near Brush Hollow Reservoir during spring and mercury concentrations were below detection in all samples, early summer of 2009. Total mercury was measured in the and total mercury concentrations ranged narrowly from 0.53 supply ditch (S6) and Brush Hollow Creek (S8) to investigate to 0.70 ng/L, with the exception of a slightly elevated con- other potential sources of mercury to the reservoir. The supply centration (1.46 ng/L) in the deepest sample from site 7B in ditch derives water from the Beaver Creek drainage, which has Pueblo Reservoir. Because concentrations were measured in headwater streams that extend into a historical mining district. unfiltered samples, the elevated total mercury concentration The Brush Hollow Creek site, which is downstream from the in the deep sample at Pueblo Reservoir in August 2008 may reservoir, was sampled because thermal springs, which could indicate a greater particulate fraction in the hypolimnetic water be a potential geologic source of mercury, exist just upstream Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs 11

Table 5. Mercury concentrations in water samples from Pueblo and Brush Hollow Reservoirs, Colorado.

[No., site number from fig. 1; MeHg, methylmercury in nanograms per liter;THg, total mercury in nanograms per liter; <, less than; --, not analyzed]

No. Site name Sample date Type MeHg THg 4B Pueblo Reservoir Site 4B 08/05/2008 Epilimnion <0.04 0.55 4B Pueblo Reservoir Site 4B 08/05/2008 Hypolimnion <.04 .67 7B Pueblo Reservoir Site 7B 08/05/2008 Epilimnion <.04 .53 7B Pueblo Reservoir Site 7B 08/05/2008 Hypolimnion <.04 1.46 S1 Brush Hollow Reservoir Site S1 08/27/2008 Epilimnion <.04 .60 S1 Brush Hollow Reservoir Site S1 08/27/2008 Hypolimnion <.04 .70

S1 Brush Hollow Reservoir Site S1 04/23/2009 Epilimnion -- .46 S1 Brush Hollow Reservoir Site S1 04/23/2009 Hypolimnion -- .49 7B Pueblo Reservoir Site Site 7B 05/19/2009 Epilimnion <.04 .41 7B Pueblo Reservoir Site Site 7B 05/19/2009 Hypolimnion <.04 .40 S1 Brush Hollow Reservoir Site S1 06/25/2009 Epilimnion .05 .60

S6 Brush Hollow Ditch 04/23/2009 Supply ditch -- 2.22 S8 Brush Hollow Creek At Hwy 50 04/23/2009 Stream -- 1.29 S7 Brush Hollow Reservoir Nr Boat Ramp 05/19/2009 Shore .16 2.84 S7 Brush Hollow Reservoir Nr Boat Ramp 06/25/2009 Shore .09 1.01 from the sampling site. Total mercury concentrations at these sediment from Pueblo Reservoir (45.8 ng/g) was nearly twice two sites were slightly higher than in the reservoir water, but that for Brush Hollow Reservoir (26.0 ng/g), and the median concentrations were still quite low, indicating inputs from methylmercury concentration in sediment from Pueblo Reser- land-use activities (mining) or geologic sources (springs) voir (0.68 ng/g) was more than 3 times that for Brush Hollow likely are small. Methylmercury concentrations are not avail- Reservoir (0.19 ng/g). Mercury concentrations in bottom sedi- able for any of the April 2009 samples due to a miscommu- ment from both study reservoirs were similar to those reported nication with the laboratory. Two samples of reservoir water for stream sediments collected from across the U.S. (Scudder also were collected near the shoreline (site S7) in May and and others, 2009). Stream sediments in unmined basins had a June 2009. Both of these samples had higher total mercury median total mercury concentration of 30.3 ng/g with 75 per- concentrations than the midlake samples from site S1 and, cent of sites less than 80 ng/g and a median methylmercury more importantly, both samples had detectable methylmercury concentration of 0.51 ng/g with 75 percent of sites less than concentrations, which accounted for as much as 8 percent of 2 ng/g (Scudder and others, 2009). By contrast, concentra- the total mercury. Several explanations are possible for higher tions were substantially higher in a historical gold-mining methylmercury in these samples, although the limited amount area in California, where the median total mercury concentra- of data makes interpretation difficult. Because the samples tion in stream sediment was around 1,000 ng/g (Alpers and were collected near the shoreline and were unfiltered, elevated others, 2005). The ratios of methylmercury to total mercury methylmercury could result from a greater particulate fraction concentrations in Brush Hollow and Pueblo bottom sediments caused by suspension of sediments by wave action in shal- ranged from 0.006 to 0.021, with the exception of the sample low areas. Another possibility is that higher methylmercury in from site 4B, which had a higher ratio of 0.055 owing to a these samples might indicate enhanced production of methyl- comparatively low total mercury concentration (table 6). The mercury in sediments in near-shore areas. sample from site 4B was collected at the shoreline of Pueblo Reservoir and might have a greater influence from local geo- Mercury in Reservoir Bottom Sediment logic inputs compared to sediments collected on the reservoir bottom. Total mercury concentrations in bottom sediment ranged Patterns of mercury concentrations in bottom sediment in from 23.1 to 52.2 ng/g and methylmercury concentrations the two reservoirs differed as a function of water depth at the ranged from 0.14 to 1.30 ng/g in the samples from August sampling site (fig. 5). In Brush Hollow Reservoir, methylmer- 2008 (table 6). Despite higher fish-tissue mercury in Brush cury concentrations increase with increasing water depth with Hollow Reservoir, the median total mercury concentration in a maximum value at the deepest point of the reservoir near the 12 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Table 6. Mercury concentrations in reservoir-bottom sediments from Pueblo and Brush Hollow Reservoirs, Colorado.

[No., site number in figure 1; ng/g, nanograms per gram dry weight]

Ratio of Organic Collect Methylmercury Total mercury No. Site name methylmercury carbon date (ng/g) (ng/g) to total mercury (percent) 7B Pueblo Reservoir 7B 08/06/2008 0.61 49.1 0.012 10.5 6C Pueblo Reservoir 6C 08/06/2008 .53 45.8 .012 9.9 5C Pueblo Reservoir 5C 08/06/2008 .68 40.0 .017 9.7 4B Pueblo Reservoir 4B 08/06/2008 1.30 23.6 .055 9.4 3B Pueblo Reservoir 3C 08/06/2008 1.10 52.2 .021 10.5 Median concentration .68 45.8 .017 9.9

S1 Brush Hollow Reservoir S1 08/27/2008 .54 34.5 .016 10.8 S2 Brush Hollow Reservoir S2 08/27/2008 .32 34.9 .009 11.3 S3 Brush Hollow Reservoir S3 08/27/2008 .16 26.0 .006 9.6 S4 Brush Hollow Reservoir S4 08/27/2008 .14 24.6 .006 9.7 S5 Brush Hollow Reservoir S5 08/27/2008 .19 23.1 .008 9.0 Median concentration .19 26.0 .008 9.7

Brush Hollow Reservoir Pueblo Reservoir 0.6

1.2

0.4 0.8

0.2 0.4 METHYLMERCURY, IN METHYLMERCURY, IN METHYLMERCURY, NANOGRAMS PER GRAM NANOGRAMS PER GRAM

0.0 0.0

0.06 0.015

0.04 0.010

0.005 0.02 TO TOTAL MERCURY TO TOTAL TO TOTAL MERCURY TO TOTAL RATIO OF METHYLMERCURY RATIO RATIO OF METHYLMERCURY RATIO

0.000 0.00 0 5 10 15 20 0 30 60 90 120 RESERVOIR DEPTH, IN FEET RESERVOIR DEPTH, IN FEET

Figure 5. Relation of methylmercury and the ratio of methylmercury to total mercury concentrations with water depth for reservoir-bottom sediments collected during August 2008 in Pueblo and Brush Hollow Reservoirs. Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs 13 dam. The ratio of methylmercury to total mercury showed a Methylmercury accounted for about 61 percent of total mer- similar pattern indicating the anoxic conditions in deeper parts cury in zooplankton samples from Brush Hollow Reservoir of the reservoir might enhance methylmercury production in and 50 percent in samples from Pueblo Reservoir collected sediments. Sediment in Pueblo Reservoir showed the oppo- during late summer. Despite higher mercury in fish at Brush site pattern with the lowest methylmercury concentrations Hollow Reservoir, mercury concentrations in zooplankton and ratios of methylmercury to total mercury in the deepest were higher in Pueblo Reservoir at the time of sampling, part of the reservoir where oxygen concentrations are most although the percentage of methylmercury was slightly higher depleted (fig. 5). The opposite pattern might result from the at Brush Hollow Reservoir. The low concentrations and lack location of primary inflow sources relative to sampling sites. of substantial difference in concentrations between the reser- The main inflow to Pueblo Reservoir is from the Arkansas voirs likely reflects the timing of sampling. Large variations in River that enters at the shallow end of the reservoir, whereas zooplankton mercury concentrations have been reported for a the main inflow to Brush Hollow Reservoir is from a ditch that snowmelt-dominated reservoir in California with methylmer- enters the reservoir near the dam (fig. 1). Inputs of organically cury concentrations in spring nearly an order of magnitude bound mercury from the inflows could increase the amount of greater than during fall and winter (Stewart and others, 2008), mercury available for methylation (Wiener and others, 2006). indicating zooplankton concentrations in the study reservoirs Additionally, greater amounts of organic material in sediments might be different during other seasons of the year. deposited in areas closer to inflow sources could provide carbon for sulfate reduction, contributing to increased methylmercury production in these areas. Sediment-Incubation Experiments Measurement of methylmercury production rates is one Mercury in Zooplankton potentially effective way to reveal the influence of sediment quality across differing systems on methylmercury abundance Zooplankton samples were collected in both reservoirs (Marvin-DiPasquale and others, 2009). In this study, labora- during August 2008 to compare mercury in biota at the bottom tory experiments were conducted to compare the methylation of the food chain. Total mercury concentrations in zooplankton potential of littoral sediments from the two study reservoirs. ranged from 19.7 to 42.8 ng/g and methylmercury concentra- The experiments involved incubating sediment cores collected tions ranged from 11.7 to 19.2 ng/g (fig. 6). These values are at the shoreline with reservoir water and periodically monitor- at the low end of concentrations reported for temperate lakes ing methylmercury concentrations in the water column over in the northeastern United States (Driscoll and others, 2007). a period of several weeks. A spike of inorganic mercury-201 was added to the water to monitor the production of methylmercury during the incubations. Changes in methylmercury concentrations relative to 20 duration of the experiment in days are presented in figure 7 for the three sediment cores collected from each reservoir. 15 The sample at day 0 is the methylmercury concentration in the spiked water sample for both the mercury-201 tracer and the ambient mercury that was contained in the surface water at 10 the time of sampling. For all the incubations, the fastest rate of accumulation of ambient methylmercury occurred during METHYLMERCURY, IN METHYLMERCURY,

NANOGRAMS PER GRAM 5 the first two weeks of the experiment. For ambient methylmercury, the Brush Hollow Reservoir incubations yielded considerably more methylmercury (about 4 to 10 times) than those 0 0 10 20 30 40 50 of the Pueblo Reservoir (fig. 7), although appreciable vari- TOTAL MERCURY, IN ability is exhibited among individual Brush Hollow samples. NANOGRAMS PER GRAM The initial increase in aqueous methylmercury indicates that either methylmercury contained in the sediment cores was EXPLANATION released to the overlying solution or that wetting stimulated Pueblo Reservoir methylmercury production in the sediment, which then dif- Brush Hollow Reservoir fused into the water column. Like ambient methylmercury, methylmercury-201 showed a rapid appearance during the first two weeks of the experiment, especially for the Brush Hol- Figure 6. Relation between total mercury low samples (see bottom panels of fig. 7), which have about concentrations and methylmercury concentrations in 2 times greater methylmercury-201 at any particular time zooplankton samples collected from Pueblo and Brush during the experiment. The appearance of methylmercury-201 Hollow Reservoirs during August 2008. in solution is direct evidence that these sediments support the 14 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Brush Hollow Reservoir Pueblo Reservoir

BH1 2.5 2.5 PR1 BH2 PR2 2.0 2.0 BH3 PR3 1.5 1.5

1.0 1.0 METHYLMERCURY, METHYLMERCURY, 0.5 METHYLMERCURY, 0.5 IN NANOGRAMS PER LITER IN NANOGRAMS PER LITER 0.0 0.0

0 10 20 30 40 50 0 10 20 30 40 50 60

0.2 0.2

0.1 0.1 METHYLMERCURY-201, METHYLMERCURY-201, METHYLMERCURY-201, METHYLMERCURY-201, IN NANOGRAMS PER LITER IN NANOGRAMS PER LITER 0.0 0.0

0 10 20 30 40 50 0 10 20 30 40 50 60 TIME, IN DAYS TIME, IN DAYS

Figure 7. Methylmercury concentration relative to duration of the experiment in days for the sediment-incubation experiments. methylation process because the isotopic tracer was added as inorganic mercury. For the most part, the methylmercury-201 production profiles mirrored those of the ambient methylmercury, supporting the notion that the observed ambient 3 methylmercury production profiles reflect newly produced BH3 methylmercury, as opposed to previously existing methylmercury contained in the sediment at the time of sampling. Even though the sediments were exposed (above the water level) 2 when they were collected, some methylmercury was produced once they were rewetted, indicating that drying of sediments BH1 r2 = 0.86 did not inhibit methylation upon rewetting. In fact, other stud- 1 ies have shown that drying and rewetting of soils may actually stimulate methylation (Gilmour, 2003). Based on the incuba- IN NANOGRAMS PER GRAM PR3 METHYLMERCURY CONCENTRATION, METHYLMERCURY PR2 tion results, the near-shore sediments from Brush Hollow Res- PR1 BH2 ervoir appear to have a substantially greater potential (about 0 2–10 times) to yield methylmercury to the water column than 0 4 8 12 those from Pueblo Reservoir when reflooding occurs. The PERCENT ORGANIC CARBON difference in experimental results from these two reservoirs seems largely to be explained by the greater organic carbon EXPLANATION content of the sediments from Brush Hollow. Overall, a strong Pueblo Reservoir positive correlation was found between the maximum methyl- Brush Hollow Reservoir mercury concentration observed from each experiment and the organic carbon content of the sediment (r2 = 0.86) (fig. 8). It Figure 8. Relation between maximum methylmercury is interesting to note that all the experiments (with the excep- concentration in water and organic-carbon content of tion of core BH1) showed decreasing or steady methylmercury cores in sediment-incubation experiments. Factors Affecting Bioaccumulation of Mercury 15 concentrations during the latter part of the experiment. These methylation (Wiener and others, 2003). In low-sulfate systems, time profiles indicate that the net methylmercury production added sulfate will stimulate methylation; however, in systems declined and(or) ceased after about two weeks or that de- with high sulfate, the accumulation of sulfide (the byproduct methylation of methylmercury accelerated over the time span of sulfate reduction) can inhibit methylmercury production of the experiment and eventually exceeded production after (Gilmour and others, 1992; Benoit and others, 1999). Gilmour about two weeks. (2003) reported that, in the Everglades, sulfate concentrations between 1 and 10 mg/L were optimal for methylation and concentrations above 50 mg/L inhibited methylmercury production due to sulfide accumulation in pore water. The high Factors Affecting Bioaccumulation of sulfate concentrations (36.7 to 86.8 mg/L; table 4) in water in Mercury the study reservoirs indicate that high sulfate conditions are not likely a limiting factor in methylmercury production in the One factor that can cause elevated mercury concentra- study reservoirs. tions in fish is a larger mercury input from point-source In temperate lakes, the strongest correlation with fish discharges related to mining or urban activities or from mercury typically is with DOC, which often is attributed elevated rates of atmospheric deposition (Evers and others, to runoff from adjacent wetlands, as they are active areas 2007). No industrial point sources or large urban centers exist of methylmercury production (Wiener and others, 2006). in either of the reservoir watersheds; however, both drainage However, neither study reservoir has adjoining wetlands, areas are affected by historical mining activities. Mining in the indicating that DOC and methylmercury inputs from wetland headwaters of Beaver Creek primarily was for gold; how- sources are not influencing factors. DOC concentrations in ever, due to the nature of the ore, the gold was not reclaimed reservoir water samples were slightly higher in Brush Hollow by amalgamation with mercury (http://ccvgoldmining.com/ Reservoir than Pueblo Reservoir during August 2008 (table 4), Geology/geology.html, accessed January 2010). The headwa- perhaps reflecting differences in primary production. DOC ters area of the Arkansas River primarily was mined for silver can enhance the solubility and production of methylmercury and, aside from one large placer deposit near Leadville, gold but also can bind with methylmercury and reduce its bioavail- mining activities were limited to small placer deposits and ability (Driscoll and others, 2007; Gorski and others, 2008). prospects (Emmons and others, 1927; Parker, 1974). These The relation between DOC and methylmercury in lakes is descriptions indicate historical mining is not a major source of not always clear, especially at low concentrations (<5 mg/L). mercury to the study reservoirs. In addition, mercury concen- Wisconsin seepage lakes (Watras and others, 1995) show trations in bottom sediments collected at each reservoir were little relation between DOC and methylmercury in low-DOC relatively low (fig. 5), particularly when compared to sediment lakes; whereas, high-elevation lakes across the Western United concentrations in mining-affected areas (Alpers and others, Stated reveal a positive relation (Krabbenhoft and others, 2005). Thermal springs along the Arkansas River valley also 2002). This general lack of concurrence prevents us from con- were ruled out as a potential mercury source on the basis of cluding that accumulation of DOC is the main factor control- low mercury concentration in Brush Hollow Creek at site S8 ling differences in fish-tissue mercury concentrations between (table 5). the study reservoirs and points to the need for more work to Given that geologic and point sources of mercury prob- answer this question. ably are minor, atmospheric deposition appears to be the domi- Nutrient enrichment of lakes and reservoirs also has nant source of mercury accumulating in the study reservoirs. been correlated with lower mercury levels in fish (Chen and Due to the proximity of the reservoirs and the importance of others, 2005). This correlation is thought to occur as a result regional or even global mercury sources in controlling deposi- of algal blooms in productive water bodies, which reduce tion patterns (Selin and Jacob, 2008), current atmospheric mercury in the food chain through the process of biodilution deposition probably does not vary sufficiently to account for (Pickhardt and others, 2002). The water-quality results from differences in fish-tissue mercury. Therefore, water-quality August 2008 showed the opposite pattern, with higher fish and (or) landscape characteristics are likely the main factors mercury in the more productive reservoir (Brush Hollow), controlling bioaccumulation of mercury in the study reservoirs indicating that nutrient enrichment is not a major factor unless (Wiener and others, 2006). plankton growth and nutrient inputs are vastly different during Recent studies have identified several water-quality other times of the year. A previous study of Pueblo Reservoir characteristics associated with elevated fish mercury, most showed that phytoplankton production peaked during mid- to notably pH, alkalinity, sulfate, and DOC (Chen and others, late summer (Lewis and Edelmann, 1994), indicating that the 2005). In the Northeastern United States, lakes with low pH August results are representative of trophic conditions during (less than 6.0) and low alkalinity (less than 5 mg/L) tend to the growing season. have elevated fish-tissue mercury levels (Driscoll and others, The potential for methylmercury production may be 2007). This does not fit the pattern for the study reservoirs, increased in reservoirs and lakes that undergo stratification and both of which have pH values above 8 and alkalinities over development of anoxic conditions in the hypolimnion (Cana- 70 mg/L. Sulfate has a more complicated response on mercury van and others, 2000). Buildup of methylmercury in anoxic 16 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09 layers can occur as a result of transformation of inorganic mer- to 20 to 29 percent for Pueblo Reservoir (fig. 9). These results cury by sulfate-reducing bacteria living in the water column fit the pattern of higher fish mercury concentrations in Brush or can diffuse from bottom sediments to the overlying water Hollow Reservoir with a greater area of exposed shoreline column (Watras and others, 2005). Anoxic conditions in the resulting from larger annual water-level fluctuations compared hypolimnion may also decrease rates of demethylation result- to Pueblo Reservoir. ing in a net increase in methylmercury concentrations in the Another factor that may affect methylmercury production water column (Canavan and others, 2000). In the study reser- in reflooded areas is the nature of the littoral sediments. Steep- voirs, anoxic conditions developed in the hypolimnion in mid- sided reservoirs with organic-poor substrates can be expected summer, with dissolved oxygen concentrations below 4 mg/L to display less efficient methylmercury production than in the deepest areas of the reservoirs (fig. 4). Methylmercury reservoirs with wide basins and large littoral areas with more concentrations in hypolimnetic water were not detectable even organic matter (Evers and others, 2007). Shallow sediments in during late summer (table 5), indicating stratification was not Brush Hollow Reservoir are rich in organic matter, particularly a major factor in methylmercury production or availability in at the north end of the reservoir where macrophytes are pres- either reservoir. The reason for low concentrations of methyl- ent (fig. 10A). Moreover, because drawdown typically occurs mercury in the hypolimnion is not clear but could be related by July (fig. 2), terrestrial plants have ample opportunity to sediment characteristics. In high-productivity reservoirs during the growing season to revegetate exposed littoral areas or lakes, for example, sediments may have higher levels of (note grassy areas in figure 10A–B). By contrast, the shoreline organic matter, which can bind inorganic mercury thus reduc- of Pueblo Reservoir is rocky with little organic matter and lim- ing the amount available for methylation (Hammerschmidt ited vegetation (fig. 10C–D). In addition, water levels typically and Fitzgerald, 2004). Methylmercury production also can reach a minimum after the growing season, providing less be inhibited in sediments with high levels of sulfide resulting opportunity for regrowth of plants. The importance of organic from high rates of sulfate reduction (Wiener and others, 2003). content of littoral sediments for methylmercury production is Given that water-quality characteristics do not appear supported by the results from the sediment-incubation experi- to provide a clear explanation for differences in fish mer- ments (fig. 7). For littoral sediments from Pueblo Reservoir, cury, reservoir characteristics or water levels or both, may methylmercury production was uniformly low in all three be plausible factors. The occurrence of elevated fish-tissue incubations, whereas two of the sediment samples from Brush mercury concentrations in newly created reservoirs due to the Hollow Reservoir produced considerably higher methyl initial flooding of terrestrial soils has been well established mercury concentrations. These simple experiments seem to (Bodaly and others, 2007); however, a similar response has support the hypothesis that the more organic-rich sediments in been observed for water bodies that are subject to periodic Brush Hollow Reservoir have a greater potential for methyl- water-level fluctuations (Sorensen and others, 2005; Selch and mercury production; however, sediment-water interactions and others, 2007). One explanation is that declining water levels redox conditions in the laboratory may be very different than allow the growth of vegetation on exposed littoral areas, which conditions in the field and the results need to be interpreted then become a new carbon source when the sediments are with some caution. More extensive temporal and spatial stud- reflooded, promoting increased microbial activity and methyl- ies of methylmercury in reservoir and sediment-pore water are mercury production. An alternate explanation is that drying needed to better understand the role of water-level manipu- of soils and sediments results in oxidation of reduced sulfur lations on bioaccumulation of mercury in these Colorado to sulfate which, when rewetted, stimulates sulfate-reducing reservoirs. bacteria and methylmercury production (Gilmour, 2003). Although both study reservoirs experience annual fluctua- Pueblo Brush Hollow tions in storage, the overall change typically is much greater in Brush Hollow Reservoir than Pueblo Reservoir (fig. 2). 60 Moreover, due to the shallow depth of Brush Hollow Reser- voir, fluctuations in water level likely result in proportionally larger areas of exposed shoreline compared to Pueblo Reser- 40 voir, which is deep and steep sided. To evaluate this possible factor, the change in reservoir surface area (SA) resulting from annual water-level fluctuations was estimated by using

PERCENT CHANGE IN 20

Landsat 5 satellite imagery (http://landsat.usgs.gov/, accessed AREA RESERVOIR SURFACE March 2010) and tools in a geographic information system (GIS). Percent change in SA was computed as SA at maximum 0 stage minus SA at the previous minimum stage divided by the 2002 2003 2004 SA at maximum stage. Percent SA change was computed for each of the 3 years prior to the CDPHE fish sampling, which Figure 9. Percent change in reservoir surface area between occurred in October 2004. In Brush Hollow Reservoir, SA annual minimum and maximum water level in Pueblo and Brush changed by 54 to 67 percent over the 3-year period compared Hollow Reservoirs during 2002–2004. Factors Affecting Bioaccumulation of Mercury 17 C–D ) typical rocky D B C A ) Submerged macrophytes and ( B grass-covered littoral zone at north end of Brush Hollow Reservoir in late summer 2008, (

A Figure 10. shoreline at Pueblo Reservoir. Tree line in ( A ) and B delineates water level at maximum storage Brush Hollow Reservoir. Tree shoreline at Pueblo Reservoir. 18 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Summary higher methylmercury production from sediments from Brush Hollow might result from the higher organic carbon content in the sediments. In 2004, the Colorado Department of Public Health and Results of the study were used to evaluate factors that can Environment (CDPHE) initiated a 5-year study to test fish affect bioaccumulation of mercury including mercury sources, tissue for mercury in water bodies across the State. Of the water chemistry, and reservoir characteristics. Atmospheric water bodies tested, nearly one-quarter had fish with mercury deposition appears to be the dominant source of mercury; concentrations that exceeded the Colorado fish-tissue criterion however, owing to the proximity of the reservoirs, deposition for the protection of human health. As a result of the tissue rates likely are similar in the two study areas. Several water- study, the U.S. Geological Survey, in cooperation with the quality characteristics associated with elevated fish mercury CDPHE, conducted a study to investigate environmental fac- (pH, alkalinity, sulfate, and dissolved organic carbon) were tors that may contribute to the bioaccumulation of mercury in evaluated but did not provide clear explanations for differ- two Front Range reservoirs. One of the reservoirs, Brush Hol- ences between the reservoirs. Nutrient enrichment can result in low Reservoir, has a fish-consumption advisory for mercury in walleye (Stizostedion vitreum), and the other, Pueblo Reser- diminished mercury accumulation in fish through biodilution; voir, does not. This report compares mercury concentrations in however, the opposite pattern was observed at the study sites water, bottom sediment, and zooplankton samples collected at with higher fish mercury in the more productive of the two the two reservoirs during 2008 and 2009, presents results from reservoirs (Brush Hollow). Development of anoxic condi- a sediment-incubation study conducted in 2009, and evaluates tions in the hypolimnion during summer also did not appear factors that may be affecting bioaccumulation of mercury in to enhance methylmercury production in either reservoir. fish. Water-level fluctuations and shoreline characteristics appear to Methylmercury concentrations in water from the two best explain differences in fish-tissue mercury concentrations study reservoirs were below detection in all August 2008 sam- between the reservoirs. Due to the shallow depth of Brush ples, and total mercury showed a very narrow range (0.53 to Hollow Reservoir and the large annual water-level fluctua- 0.70 ng/L) with the exception of a slightly elevated concentrations, proportionally larger areas of shoreline are subjected to tion (1.46 ng/L) in the deepest sample from Pueblo Reservoir. annual reflooding compared to Pueblo Reservoir. Moreover, During spring of 2009, water from the midlake sites at both macrophytes and regrowth of terrestrial vegetation likely reservoirs had low concentrations of methylmercury and total increase the organic content of Brush Hollow Reservoir mercury. Two water samples from Brush Hollow Reservoir sediments, which may stimulate methylmercury production also were collected near the shoreline in May and June 2009. in reflooded sediments. Results of a laboratory incubation Both of these samples had higher total mercury concentrations experiment are consistent with this hypothesis, showing higher than the midlake samples and had detectable methylmercury methylmercury production for the near-shore sediments from concentrations, which accounted for as much as 8 percent of Brush Hollow Reservoir than those from Pueblo Reservoir. the total mercury. Mercury concentrations in reservoir bottom sediments were similar to those reported for stream sediments from References unmined basins across the United States. Bottom-sediment total mercury concentrations ranged from 23.1 to 52.2 ng/g, and methylmercury concentrations ranged from 0.14 to Alpers, C.N., Hunerlach, M.P., May, J.T., Hothem, R.L., Tay- 1.3 ng/g. Despite higher fish-tissue mercury in Brush Hollow lor, H.E., Antweiler, R.C., De Wild, J.F., and Lawler, D.A., Reservoir, the median total mercury concentration in bottom 2005, Geochemical characterization of water, sediment, and sediment in Pueblo Reservoir was nearly twice that in Brush biota affected by mercury contamination and acidic drain- Hollow Reservoir, and the median methylmercury sediment age from historical gold mining, Greenhorn Creek, Nevada concentration in Pueblo Reservoir was more than 3 times that County, California, 1999–2001: U.S. Geological Survey in Brush Hollow Reservoir. Total mercury in zooplankton Scientific Investigations Report 2004–5251, 278 p. ranged from 19.7 to 42.8 ng/g, and methylmercury concentra- Benoit, J.M., Gilmour, C.G., Mason, R.P., and Heyes, A., tions ranged from 11.7 to 19.2 ng/g. Methylmercury accounted 1999, Sulfide controls on mercury speciation and bioavail- for about 61 percent of total mercury in Brush Hollow Reser- ability to methylating bacteria in sediment pore saters: Envi- voir samples and 50 percent in Pueblo Reservoir samples dur- ronmental Science and Technology, v. 33, p. 951–957. ing late summer. The similarity in zooplankton concentrations between sites is attributed to the timing of sampling. Mercury Bodaly, R.A., Jansen, W.A., Majewski, A.R., Fudge, R.J.P., concentrations in zooplankton during other seasons may be Strange, N.E., Derksen, A.J., and Green, D.J., 2007, Postim- very different. poundment time course of increased mercury concentrations Results of the sediment-incubation experiments showed in fish in hydroelectric reservoirs of northern Manitoba, higher methylmercury production for littoral sediments from Canada: Archives of Environmental Contamination Toxicol- Brush Hollow Reservoir compared to Pueblo Reservoir. The ogy, v. 53, no. 3, p. 379–389. References 19

Bureau of Reclamation, 1972, Pueblo Dam and Reservoir final Galloway, J.M., Ortiz, R.F., Bales, J.D., and Mau, D.P., 2008, environmental impact statement: Denver, Final Environ- Simulation of hydrodynamics and water quality in Pueblo mental Statement FES–72–16, 81 p. Reservoir, southeastern Colorado, for 1985 through 1987 and 1999 through 2002: U.S. Geological Survey Scientific Bureau of Reclamation, 1977, Pueblo Reservoir area-capacity Investigations Report 2008–5056, 87 p. tables and curves: Denver, Colorado, U.S. Department of the Interior, Fryingpan-Arkansas Project, Colorado. Gilmour, C.C., Henry, E.A., and Mitchell, R., 1992, Sulfate stimulation of mercury methylation in freshwater sediments: Canavan, C.M., Caldwell, C.A., and Bloom, N.S., 2000, Dis- Environmental Science and Technology, v. 26, charge of methylmercury-enriched hypolimnetic water from p. 2281–2287. a stratified reservoir: The Science of the Total Environment, v. 260, p. 159–170. Gilmour, C.C., 2003, Status report on the effect of water quantity and quality on methylmercury production: Carlson, R.E., 1977, A trophic state index for lakes: Limnol- Everglades Consolidated Report. Appendix 2B–2, ogy and Oceanography, v. 22, no. 2, p. 361–369. https://my.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_ sfer/portlet_prevreport/ecr2003/appendices/app2b-2.pdf, Chen, C.Y., Stemberger, R.S., Kamman, N.C., Mayes, B.M., accessed January 2010. and Folt, C.L., 2005, Patterns of mercury bioaccumulation and transfer in aquatic food webs across multi-lake studies Gorski, P.R., Armstrong, D.E., Hurley, J.P., and Krabbenhoft, in the northeast US: Ecotoxicology, v. 14, p. 135–147. D.P., 2008, Influence of natural dissolved organic carbon on the bioavailability of mercury to a freshwater alga: Environ- DeWild, J.F., Olson, M.L., and Olund, S.D., 2001, Determina- mental Pollution, v. 154, p. 116–123. tion of methyl mercury by aqueous phase ethylation fol- lowed by gas chromatographic separation with cold vapor Hammerschmidt, C.R., and Fitzgerald, W.F., 2004, Geochemi- atomic fluorescence detection: U.S. Geological Survey cal controls on the production and distribution of methyl- Open-File Report 2001–445, 14 p. mercury in near-shore marine sediments: Environmental Science and Technology, v. 38, no. 5, p. 1487–1495. DeWild, J.F., Olund, S.D., Olson, M.L, and Tate, M.T., 2004, Methods for the preparation and analysis of solids and sus- Hem, J.D., 1985, Study and interpretation of the chemical pended solids for methylmercury: U.S. Geological Survey characteristics of natural water: U.S. Geological Survey Techniques and Methods, book 5, chap. A7, 13 p. Water-Supply Paper 2254, 263 p.

Driscoll, C.T., Blette, V., Yan, C., Schofield, C.L., Munson, Kolka, R.K., Grigal, D.F., Verry, E.S., and Nater, E.A., 1999, Mercury and organic carbon relationships in streams drain- R., and Holsapple, J., 1995, The role of dissolved organic ing forested upland/peatland watersheds: Journal of Envi- carbon in the chemistry and bioavailability of mercury in ronmental Quality, v. 28, p. 766–775. remote Adirondack lakes: Water, Air, and Soil Pollution, v. 80, no. 1–4, p. 499–508. Krabbenhoft, D.P., Olson, M.L., Dewild, J.F., Clow, D.W., Striegl, R.S., Dornblaser, M.M., and VanMetre, Peter, 2002, Driscoll, C.T., Han, Y.J., Chen, C.Y., Evers, D.C., Lambert, Mercury loading and methylmercury production and cycling K.F., Holsen, T.M., Kamman, N.C., and Munson, R.K., in high-altitude lakes from the western United States: Water, 2007, Mercury contamination in forest and freshwater Air and Soil Pollution Focus, v. 2, p. 233–249. ecosystems in the northeastern United States: BioScience, v. 57, p. 17–28. Lewis, M.E., and Edelmann, P.R., 1994 Physical, chemical, and biological characteristics of Pueblo Reservoir, Colo- Emmons, S.F., Irving, J.D., and Loughlin, G.F., 1927, Geology rado, 1985–89: U.S. Geological Survey Water-Resources and ore deposits of the Leadville mining district, Colorado: Investigations Report 94–4097, 71 p. U.S. Geological Survey Professional Paper 148, 368 p. Marvin-DiPasquale, M., Lutz, M.A., Brigham, M.E., Krab- Evers, D.C., Han, Y.J., Driscoll, C.T., Kamman, N.C., Goo- benhoft, D.P., Aiken, G.R., Orem, W.H., and Hall, B.D., dale, W.M., Lambert, K.F., Holsen, T.M., Chen, C.Y., Clair, 2009, Mercury cycling in stream ecosystems. 2. Benthic T.A., and Butler, T.J., 2007, Biological mercury hotspots in methylmercury production and bed sediment pore-water the Northeastern United States and Southeastern Canada: partitioning: Environmental Science and Technology, v. 43, BioScience, v. 57, p. 1–15. p. 2726–2732.

Fishman, M.J., and Friedman, L.C., 1989, Methods for Olund, S.D., DeWild, J.F., Olson, M.L., and Tate, M.T., 2004, determination of inorganic substances in water and fluvial Methods for the preparation and analysis of solids and sediments: U.S. Geological Survey Techniques of Water- suspended solids for total mercury: U.S. Geological Survey Resources Investigations, book 5, chap. A1, 545 p. Techniques and Methods, book 5, chap. A8, 23 p. 20 Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado, 2008–09

Parker, B.H., Jr., 1974, Gold placers of Colorado: Golden, Stewart, A.R., Saiki, M.K., Kuwabara, J.S., Alpers, C.N., Quarterly of the Colorado School of Mines, v. 69, no. 3, Marvin-DiPasquale, M., and Krabbenhoft, D.P., 2008, Influ- 268 p. ence of plankton mercury dynamics and trophic pathways on mercury concentrations of top predator fish of a mining- Pickhardt, P.C., Folt, C.L., Chen, C.Y., Klaue, B., and Blum, impacted reservoir: Canadian Journal of Fisheries and J.D., 2002, Algal blooms reduce the uptake of toxic Aquatic Sciences, v. 65, no. 11, p. 2351–2366. methylmercury in freshwater food webs: Proceedings of the National Academy of Sciences, v. 99, p. 4419–4423. U.S. Environmental Protection Agency, 1997, Mercury study report to Congress: United States Environmental Protection Scott, G.R., Taylor, R.B., Epis, R.C., and Wobus, R.A., Agency, EPA-452–97–003, Office of Air and Radiation, 1978, Geologic map of the Pueblo 1 degree x 2 degrees Washington, D.C., 1,980 p. quadrangle, south-central Colorado: U.S. Geological Survey Miscellaneous Investigations Series Map I–1022, U.S. Environmental Protection Agency, 2002, Method scale: 1:250,000. 1631 Revision E—Mercury in water by oxidation, purge, and trap, and cold vapor atomic fluorescence spectrom- Scudder, B.C., Chasar, L.C., Wentz, D.A., Bauch, N.J., etry: United States Environmental Protection Agency, Brigham, M.E., Moran, P.W., and Krabbenhoft, D.P., 2009, Mercury in fish, bed sediment, and water from streams EPA–821–R–02–019, 38 p. across the United States, 1998–2005: U.S. Geological Sur- Watras, C.J., Morrison, K.A., Host, K., and Bloom, N.S., vey Scientific Investigations Report 2009–5109, 75 p. 1995, Concentration of mercury species in relationship to Selch, T.M., Hoagstrom, C.W., Weimer, E.J., Duehr, J.P., and other site-specific factors in the surface waters of northern Chipps, S.R., 2007, Influence of fluctuating water levels Wisconsin lakes: Limnology and Oceanography, v. 40, on mercury concentrations in adult walleye: Bulletin of p. 556–565. Environmental Contamination and Toxicology, v. 79, no. 1, Watras, C.J., Morrison, K.A., Kent, A., Price, N., Regnell, p. 36–40. O., Eckley, C., Hintelmann, H., and Hubacher, T., 2005, Selin, N.E., and Jacob, D.J., 2008, Seasonal and spatial Sources of methylmercury to a wetland-dominated lake in patterns of mercury wet deposition in the United States— northern Wisconsin: Environmental Science and Technol- Constraints on the contribution from North American ogy, v. 39, no. 13, p. 4747–4758. anthropogenic sources: Atmospheric Environment, v. 42, Wiener, J.G., Krabbenhoft, D.P., Heinz, G.H., and Scheuham- p. 5193–5204. mer, A.M., 2003, Ecotoxicology of mercury, Chapter 16, in Sorensen, J.A., Kallemeyn, L.W., and Sydor, M., 2005, Hoffman, D.J., Rattner, B.A., Burton, G.A., Jr., and Cairns, Relationship between mercury accumulation in J., Jr., eds., Handbook of Ecotoxicology, 2d ed.: Boca young-of-the-year yellow perch and water-level Raton, Florida, CRC Press, p. 407–461. fluctuations: Environmental Science and Technology, Wiener, J.G., Knights, B.C., Sandheinrich, M.B., Jeremiason, v. 39, p. 9237–9243. J.D., Brigham, M.E., Engstrom, D.R., Woodruff, L.G., Can- St. Louis, V.L., Rudd, J.W.M., Kelly, C.A., Beaty, K.G., non, W.F., and Balogh, S.J., 2006, Mercury in soils, lakes, Bloom, N.S., and Flett, R.J., 1994, Importance of wetlands and fish in Voyageurs National Park (Minnesota)—Impor- as sources of methylmercury to boreal forest ecosystems: tance of atmospheric deposition and ecosystem factors: Canadian Journal of Fisheries and Aquatic Science, v. 51, Environmental Science and Technology, v. 40, no. 20, p. 1065–1076. p. 6261–6268.

Publishing support provided by: Denver Publishing Service Center

For more information concerning this publication, contact: Director, USGS Colorado Water Science Center Box 25046, Mail Stop 415 Denver, CO 80225 (303) 236-4882

Or visit the Colorado Water Science Center Web site at: http://co.water.usgs.gov/ Mast and Krabbenhoft—Comparison of Mercury in Water, Bottom Sediment, and Zooplankton in Two Front Range Reservoirs in Colorado—SIR 2010–5037